The SWI/SNF complex in cancer - biology, biomarkers and therapy

Abstract

Cancer genome-sequencing studies have revealed a remarkably high prevalence of mutations in genes encoding subunits of the SWI/SNF chromatin-remodelling complexes, with nearly 25% of all cancers harbouring aberrations in one or more of these genes. A role for such aberrations in tumorigenesis is evidenced by cancer predisposition in both carriers of germline loss-of-function mutations and genetically engineered mouse models with inactivation of any of several SWI/SNF subunits. Whereas many of the most frequently mutated oncogenes and tumour-suppressor genes have been studied for several decades, the cancer-promoting role of mutations in SWI/SNF genes has been recognized only more recently, and thus comparatively less is known about these alterations. Consequently, increasing research interest is being focused on understanding the prognostic and, in particular, the potential therapeutic implications of mutations in genes encoding SWI/SNF subunits. Herein, we review the burgeoning data on the mechanisms by which mutations affecting SWI/SNF complexes promote cancer and describe promising emerging opportunities for targeted therapy, including immunotherapy with immune-checkpoint inhibitors, presented by these mutations. We also highlight ongoing clinical trials open specifically to patients with cancers harbouring mutations in certain SWI/SNF genes.

Fig. 1|. Function of SWI/SNF chromatin-remodelling complexes.

Illustration of SWI/SNF complex subfamilies and their genomic localization with respect to gene promoters and enhancers in nonmalignant cells. SWI/SNF complexes frequently localize at sites marked by histone H3 lysine 27 acetylation (H3K27ac), which is associated with active transcription, and cooperate with transcription factors to establish an open chromatin state^,. This activity can be opposed by that of the Polycomb repressor complexes (PRCs), particularly PRC2 that places the repressive H3K27 trimethylation (H3K27me3) mark via its enzymatic subunit, enhancer of Zeste homologue 2 (EZH2)^,. Canonical BAF (cBAF), polybromo-associated BAF (PBAF) and the most recently discovered non-canonical BAF (ncBAF, also known as GLTSCR1-containing BAF (GBAF)) are the three major SWI/SNF complex subfamilies^,,. cBAF activity might occur most strongly at enhancers, whereas PBAF and ncBAF are reported to be enriched at promoters, although also bind to some enhancers^,,,. Understanding of the distinct functions of these three SWI/SNF subfamilies is limited and requires further study.

Fig. 2. Frequency and pattern of SWI/SNF subunit mutations across human cancers.

The heatmap depicts the frequency of non-synonymous mutations and deletions in select genes encoding components of SWI/SNF complexes across cancer types. Overall, the figure depicts the high prevalence of mutations affecting nine SWI/SNF subunits and the context-specificity of these mutations, with most being highly enriched in certain paediatric and adult malignancies. ARID1A is the most frequently mutated SWI/SNF complex gene, followed by SMARCA4 and PBRM1. The heatmap was compiled using data generated by the The Cancer Genome Atlas (TCGA) Research Network, accessible through cBioPortal, as well as datasets sourced from various publications^–,,,,. The white background color indicates tumors in which less than 2.5% of tumors had mutations in the subunit.

Fig. 3. Translational science of cancers with SWI/SNF complex aberrations.

Illustration of reported vulnerabilities of cancers with loss-of-function mutations in SWI/SNF-complex genes, depicting both therapeutic opportunities supported only by preclinical evidence and treatments currently being evaluated in ongoing clinical trials (TABLE 1). The therapeutic targets include: residual SWI/SNF complexes^,,,; Polycomb repressive complex 2 (PRC2), mainly its enzymatic subunit, enhancer of Zeste homologue 2 (EZH2), and predominantly in SMARCB1-mutant or SMARCA4-mutant^,; components of the DNA damage repair pathway, in particular, poly(ADP-ribose) polymerase (PARP) and ATR in ARID1A-mutant cancers^,; and receptor tyrosine kinases (RTKs) in several cancers enriched for mutations in SWI/SNF-complex genes, in a context-specific manner^,,,. Targeting of Aurora A or CDK4/6 (cell-cyle kinases), MDM2 (a negative regulator of the tumour suppressor p53), autophagy or the proteasome could also be of potential therapeutic benefit in patients with cancers harbouring particular SWI/SNF-complex abberations^,,,. Additionally, mutations in several genes encoding SWI/SNF-complex subunits have been associated with sensitivity to immune-checkpoint inhibitors targeting programmed cell death 1 (PD-1) or programmed cell death 1 ligand 1 (PD-L1)^,–,,. ARID1A, AT-rich interactive domain-containing protein 1A; bromodomain-containing protein 9, BRD9; FGFR1, fibroblast growth factor receptor 1; HDACs, histone deacetylases; PDGFR, platelet-derived growth factor receptor; SMARCA2/4, SWI/SNF-related matrix-associated actin-dependent regulator of chromatin subfamily A member 2 or 4.